JP6918439B2 - Measuring method - Google Patents

Description

The present invention relates to a method of registering a measurement point on an object (body), the measurement point being recorded along a trajectory on the surface of the object by a measurement beam.

Various methods and devices that can be used to measure the eye are known in ophthalmology. As an example, optical interferometry is suitable as a non-invasive, non-contact, precise process. In addition, there are several other processes such as ultrasonic biometrics, (platide) corneal topography, Shine Fluke cameras, split microscopes and the like. These tests on the eye are usually performed using an interferometer. As an example, an OCT (optical coherence tomography) scanner is used as an interferometer, scanning the eye laterally point by point, and during the process, the stray light profile of the eye along the light beam. Record the (axial profile).

In Non-Patent Document 1, Ryan P. McNabb discloses a scanning method capable of recording ocular topography using DSOCT (distributed scanning OCT). Here, eye movement is established by SDOCT (spectral domain OCT) and is included in the measurement. The eye is scanned along a trajectory covering a circular disc in one projection. The locus begins at the edge region and extends along the first straight line through the center of the circle. The measurement beam is then deflected in the opposing edge region so that the measurement beam extends through the center point of the circle back along a second straight line offset by 90 degrees from the first straight line. And once again, it is deflected in the same direction at the edge region. This method is repeated until the starting point is reached again.

In Non-Patent Document 2, Karol Karnowski et al. Disclose a scanning method in which a measurement beam is displaced along straight lines arranged at right angles to each other, that is, along a grid.

Patent Document 1 discloses an optical system for measuring an eye using OCT. First, the spiral form, and second, the lines extending in parallel are disclosed as trajectories.

Patent Document 2 further discloses a concentric ring, a grid, and a parallel line as a locus form that the measurement beam follows.

Known methods have the disadvantage that the measurements take a relatively long time. Due to the relatively long measurement time, there is an increased risk that the measurement results will be affected by the movement of the patient's eyes, and as a result, only inaccurate measurement results can be obtained.

The essential requirements of the measurement, especially when measuring the eye, are, first of all, a short total measurement time and high resolution, i.e., large coverage of the surface being measured.

European Patent Application Publication No. 1975550 European Patent Application Publication No. 2417903

"Distributed scanning volumetric SCOCT for motion corrected corneal biometry" (1 September 2012, volume 3, number 9, Biomedical Optics Express) Biomedical Optics Express, 1 September 2012, volume 2, no. 9

An object of the present invention is to develop a method that is part of the technical field mentioned at the beginning for recording measurement points on an object, which can be carried out particularly at high speed and with high resolution.

The solution of interest is defined by the characteristics of claim 1. According to the present invention, the minimum radius of curvature of the locus is at least 1/7, preferably at least 1/5, and particularly preferably at least 1/3 of the radius of the circumference of the surface.

The present invention is further based on a device that implements the method. Therefore, the device preferably comprises one of the measuring devices described below, preferably an interferometer, particularly preferably an OCT device. The measuring device preferably comprises a controller that allows guidance of the measuring beam according to the method described above. Such controllers are well known to those of skill in the art.

Contents of the Invention In principle, the measurement method is based on length measurement, especially axial length measurement in the Z direction (see below), particularly preferably an axial scattering profile along the beam direction (so-called A scan). .. However, measurement methods can also be made available as different direct or indirect measurements. More specifically, flight time measurements using, for example, lasers or acoustically by echo solder can be provided as indirect measurements. The length can be determined directly by the confocal method or the interferometry. As an example, OCT can be used as an interference measurement method. Those skilled in the art also know more suitable measurement techniques.

In the following, the X, Y, and Z coordinates are used as a three-dimensional orthogonal system. The locus is preferably considered to be along the XY plane, and the measurement beam has a Z direction in at least one measurement direction, especially in the direction of the center point of the circumference. From this position, the measurement beam can move in parallel, i.e., can maintain the Z direction, or else, around one angle in the X direction and a second in the Y direction. Can swivel around the angle of. Moreover, the change in position can also be achieved by a combination of these variants. Unless otherwise stated, a preferred first variant is considered below, which is not a limitation on one of the variants. Therefore, it is possible to provide a telecentric optical unit capable of performing beam offsets in parallel. The length measurement is related to the Z-axis or the direction of the measurement beam, and changes in the index of refraction along the measurement axis may result in a refraction effect, which is recorded as an axial scattering profile. Methods for geometric correction and measurement length conversion are well known to those of skill in the art.

The trajectories described in more detail below are each understood as planar projections on the XY plane that intersect the measured object or have a small distance from the measured object. In practice, the trajectory may deviate from this description depending on the object being measured. As an example, if the points are recorded along the trajectory by the measurement beam at regular intervals, these points will be present at a distance on the object being measured if the object is in a plane parallel to the plane projection. Just do it. However, when this trajectory is projected onto, for example, a spherical cap, time-adjacent measurement points near the center are compared to time-adjacent measurement points within the marginal region of the spherical cap. They are close to each other.

In principle, the object to be measured may be arbitrary. Depending on the embodiment of the object, i.e., depending on the substance and size of the object, or whether the object is a living organism, the measurement methods described below may be selected or adapted accordingly. Can be parameterized. However, the method is preferably to measure or determine a measurement point on the human or animal eye, particularly preferably on the human eye, which may optionally have a contact lens or an intraocular lens. used. Here, the boundary layer (of the cornea, of the lens, of the retina) of the eye is preferably measured. These boundary layers can generally be largely explained by the spherical segments within the central region. Therefore, the intersection of the measurement beam and the boundary layer is almost present on the spherical cap. The same is true when measuring an artificial body with a nearly spherical contact surface, such as a contact lens or an intraocular lens. Thus, the object to be measured preferably has the shape of a substantially spherical segment, and the trajectory actually preferably extends along the spherical cap, but the trajectory is, in each case, on a plane. Considered to be an XY projection of (see below).

When the method is applied to the eye, the curved surface is preferably a surface area, particularly a region of the cornea or a deeper layer in the eye. However, the curved surface may also be provided by an inserted contact lens or intraocular lens.

Axial profile or A scan should be understood to mean a profile of an object along the Z axis or along the beam direction. B-scan is understood to mean an axial profile along a straight cut through an object, consisting of individual measurement points and distances. Axial profiles can be combined to form a three-dimensional model of an object, especially the eye, or a cut through the object, especially the eye. However, the data may also be used in different ways to simulate, for example, beam paths, correction lenses, or the like.

The degree of coverage is the portion of the surface that is measured where the distance from any desired point to the nearest measurement point does not exceed the critical value. A degree of 100% coverage is usually required to measure ocular topography, the surface to be measured should have a diameter of at least 7.5 mm and the critical distance should be 0.5 mm. However, depending on the requirements, it is also possible to use a smaller degree of coverage or a smaller measurement surface, for example when achieving shorter measurement times. The surface and distance relate to the XY plane where the trajectory is defined.

The radius of curvature of the locus is currently determined based on the plane projection of the locus in the XY plane. If the parameterized curve f (t) = (x (t), y (t)), the radius of curvature r is defined as follows.

Unless otherwise stated, the locus is defined below with respect to the XY plane as a two-dimensional curve. However, it will be apparent to those skilled in the art that the locus in its application as a projection on an object usually constitutes a three-dimensional curve in space that is distorted with respect to the locus. However, this spatial curve is not discussed in more detail because it depends on both the trajectory morphology and the object morphology and can therefore vary over a wide range. Moreover, the locus does not necessarily have to follow the function definition in a mathematically accurate way. Preferably, the locus has a form that exists on one of the trajectories described above or one of the trajectories described above, at least for each measurement point. In this case, the function or form is only finally defined by interpolation of the measurement points. However, in practice, the measurement point can also deviate slightly from the trajectory. As an example, the mean deviation from the locus can be less than 5%, preferably less than 1%, the diameter of the circumference of the measured surface, depending on the number of measurement points and the size of the object.

By selecting the minimum radius of curvature of the locus to be at least 1/7, preferably at least 1/5, and particularly preferably at least 1/3 of the radius of the circumference of the surface over the entire course of the locus, the measurement beam is , Device parts, especially beam deflection units or scanners, can be displaced at particularly high speeds without exposure to large accelerations. For known trajectories, there is usually a problem in that fast changes in direction are required. However, fast change of direction cannot be performed at any speed. The deceleration and subsequent acceleration of the movement of the measurement beam results in a time delay, which slows down the measurement method as a whole. As a result of the time delay, the measurement results can then be substantially disturbed by the patient's movements. The longer these measurements take, the more likely the patient will blink, and the measurement method may even need to be stopped and restarted under some circumstances.

A novel trajectory with a relatively large radius of curvature with respect to the surface being measured is here for multiple measurement points so that the measurement result can be distorted to a lesser degree of movement of the object or eye. It makes it possible to carry out measurement methods that can be carried out at high speed. However, the radius of curvature is also chosen to be small enough so that the locus can have a sufficiently large degree of coverage for relatively short path lengths of the locus, so that the surface to be measured is sufficient. Can be covered.

A problem that may exist when measuring a spherical cap is that the distance to the spherical cap along the trajectory in a plane projection changes less near the center than in the edge region in the case of a regular measurement point distribution. To receive. In other words, the distance on the spherical cap is no longer constant at a constant measurement point distance in plane projection, but rather increases radially from the center of the spherical cap. However, large changes in the distance along the beam direction (Z-axis) between the sequentially recorded measurement points result in loss of signal contrast, especially in the case of coherence measurements (S. Yun et al., Opt. Expr. 12 (13). ), 2977 (2004)). Therefore, when measuring a spherical cap, for example, when measuring an eye, it is particularly advantageous that the edge region of the plane projection of the measurement point distribution has a higher measurement point density along the trajectory than the region near the center. be. What this achieves is that the change in distance along the trajectory is small enough and therefore the signal contrast remains high enough.

Therefore, the locus preferably has a large radius of curvature in the edge region of the surface to be measured. Therefore, the advantage that arises here is that smaller distance changes along the trajectory appear in the rim region, especially when measuring the sphere, so the region of the crow with the greatest gradient (ie, the rim region of the crow). Is to be able to measure more accurately.

In the variant, the region of the locus with the maximum radius of curvature may also be near the center, in the center itself, or in the intermediate region between the edge region and the center. In general, the ideal trajectory meets the requirements for measurement resolution, so the exact morphology of the outer region of the spherical cap may not be very important in some situations, and as a result, a large radius of curvature. Can rather be chosen to be closer to the center.

Preferably, at least 90%, preferably at least 95%, of the locus, particularly preferably the whole, extends within the circumference. As a result, the trajectory can be used substantially completely to record the measurement points, and as a result, the total time of measurement can be reduced.

In the variant, a relatively large portion of the trajectory path length may also be outside the surface to be measured. As an example, at least 80%, 70%, or less of the path length can be assigned to the surface to be measured. Depending on the size of the surface being measured, the path lengths present outside the surface can vary as well. Under some circumstances, it may be advantageous for a relatively large portion of the path length, eg, at least 50% or at least 75%, to be outside the surface, especially for small surfaces.

Here, the surface fits the largest surface that can be considered to determine the measurement point. Under some circumstances, different parts of the surface can be evaluated and the surface can be thought of as a combination of multiple parts.

Preferably, the locus has a maximum radius of curvature, which is particularly over a path length greater than 50% of the locus, preferably over a path length greater than 75%, particularly preferably over the entire path length of the surface. It is less than the radius of the circumference. As a result, it is possible to form a trajectory that has a relatively large radius of curvature as a whole and can therefore be traced at high speed by the measurement beam.

In a variant, the locus can also have a region with a radius of curvature greater than the circumference of the surface. In particular, the locus can also have a straight portion with an infinite radius of curvature.

Therefore, it is also necessary to provide a relatively small radius of curvature if the trajectory is present in the surface as a result of the maximum radius of curvature exceeding the radius of the circumference, thus achieving sufficient coverage. Is intended to provide a radius of curvature that is less than 50% of the radius of circumference, which in turn has a negative effect on the measurement time.

Preferably, the maximum radius of curvature is less than 99%, preferably less than 95%, more specifically less than 90% of the radius of the circumference. As a result of the entire trajectory having a smaller radius of curvature compared to the circumference, it is possible to achieve a particularly uniform trajectory, especially a uniform trajectory without relatively large changes in curvature, and therefore the measurement beam. Can be passed especially fast.

In the modified example, the maximum radius of curvature may reach or exceed the radius of the circumference, as described above.

Preferably, the curvature of the locus towards the center of the circumference increases monotonically, more specifically strictly monotonously. Larger measurement point densities are achieved as a result of the locus having a smaller curvature in the edge region of the surface to be measured compared to the region closer to the center. This is advantageous especially in the case of spherical caps, such as in the case of the eye. The reason is that the density of measurement points on the spherical cap can, as a result, be maintained sufficiently high even in the marginal region.

In the variants, the radius of curvature can also increase towards the center point of the circumference, especially if the region near the center is intended to be measured more accurately due to the morphology of the object.

Preferably, the locus has intersections that are recorded at least twice with a time delay. As a result, the movement of the measured object can be detected. Therefore, the movement of the object can be filtered from the measurement result, and more precise measurement data can be obtained.

In a variant, the locus can also have points that are only passed twice.

Preferably, the locus has at least two spaced intersections, in particular the intersections are recorded three or more times with a time delay. As a result of the two distant intersections, the detection of movement of the object can be improved. The reason is that as a result, it is possible to identify not only the magnitude of movement, but also the symptoms. Similarly, it is possible to optimize motion detection by passing the intersection three or more times, thereby detecting temporal changes in motion. In general, it is also possible that there are three or more distant intersections, and as a result, motion detection can be further optimized, and thus, in particular, motion along different directions can be detected. be.

In the modified example, it is possible to eliminate a plurality of intersections or intersections that are passed three or more times.

Preferably, at least two intersections have an intersection angle greater than 90 degrees in plane projection.

In the modified example, the crossing angle may also be simply 90 degrees.

Preferably, the measurement beam follows a locus with three or more intersections, with n intersections each on one concentric ring of the k concentric rings for ^{k * n intersections.} do. Particularly preferably, the locus has five or more intersections. The reason is that a trivial solution for two concentric rings or one ring with two intersections is possible in the case of two intersections.

Unless otherwise stated, the term intersection in each case is considered to be a simple intersection that is passed exactly twice by the measurement beam. The locus can have one intersection at the center of the circumference, which results in the appearance of k ^* n + 1 intersections overall. However, in this case, the center point of the circumference is not considered to be one of the concentric rings, especially since one of the concentric rings generally does not exist as a simple intersection. ..

In the first case, the locus has exactly four intersections on one ring or two intersections on two concentric rings, in addition to the central intersection (as mentioned earlier, of the circumference). Intersections that can be in the center are not counted here).

In a further example, the locus has, for example, 40 intersections on 5 concentric rings, thereby resting 8 intersections on each ring.

Any number of intersections can be divided into two groups of factors according to the principle of prime factorization, first assigned to n intersections per ring and then to k rings. It is clear to those skilled in the art that it can be done. Therefore, 40 intersections are divided between two rings with 20 intersections in each case, four rings with 10 intersections in each case, and eight rings with five intersections in each case. The number of intersections and rings can be exchanged, with the necessary changes.

In the modified example, the intersections of the trajectories can be arranged differently. As an example, n intersections and m intersections, respectively, can be present on every other (alternating) ring. Here, for example, n can be a multiple of m, and one intersection is passed three or more times.

Preferably, the distance between two adjacent concentric rings with increasing radii decreases between the three concentric rings with maximum radii. As a result, the ring is clogged in the direction of the edge region, and as a result, the intersections are also closer together. Therefore, the measurement point density increases in the edge region, and the measurement accuracy can be improved especially in the case of a ball object such as an eye. At the same time, eye movements can be detected more accurately. The reason is that the distance in the Z direction also changes more significantly in the marginal region of the eye in the case of lateral changes in the distance in the XY plane, and therefore the distance in the Z direction is used to detect the location of the eye. Because it can be done.

In a variant, the rings can also have constant spacing, or the spacing can be reduced towards the center of the circumference.

Preferably, the intersection between the locus and each of the other concentric rings lies on a straight line oriented in the radial direction. As a result, the intersections can be distributed in an ideal way across the surface being measured, especially if the rings have a shorter distance from each other than the two adjacent intersections on the individual rings. In the case of a certain number of n intersections for each ring, in this case, there are offsets of adjacent rings with respect to the intersections by an angle of 360 / 2n degrees.

In the modified example, the ring can also be offset by an angle ^{of 360 / k * n degrees.} Here, k> 2. Moreover, it is possible when k <2, especially when k = 1, so that when k = 1, there is one radially oriented straight line recording one intersection on each ring. Record the intersection. Finally, the intersections can be similarly arranged differently or chaotically in multiple rings.

Preferably, the measurement beam is displaced along the projection of the trajectory at a constant angular velocity. The result is a trajectory that is particularly easy to parameterize and analyze. Moreover, its trajectory is easy to calibrate.

In the modified example, the locus can also be traced by the measurement beam, not at a constant angular velocity, but rather at, for example, a constant path velocity, or a velocity that is neither a constant path velocity nor a constant angular velocity.

Preferably, the measurement points are recorded by a spectral region OCT (SD-OCT) or a sweep source OCT (SS-OCT), which preferably has a time-constant frequency. Methods of using OCT are well established in ophthalmology. The reason is that these methods are used in scatterable objects such as the eye and can have a relatively high penetration depth and at the same time have a high axial resolution.

Different optical frequencies are used in SD-OCT. The light is dispersed and analyzed by a CCD or CMOS sensor. As a result, the total measurement depth can be obtained in a single measurement. In contrast, in SS-OCT, the optical frequency is adjusted periodically and the interference signal is measured in a time-resolved manner. These techniques are well known to those of skill in the art.

In variants, it is also possible to use point-based length measurements or profile measurements (recording A scans) (see interferometer), for example, other techniques for measuring flight time with a laser or similar. be.

Preferably, the locus is given by the loops, the adjacent loops intersect, and in particular the two adjacent loops each have intersections located on circles that are preferably concentric with the circumference of the surface being measured. The concentric circles preferably have a radius smaller than the circumferential radius. The loops preferably all have a common intersection preferably located in the center of the circumference. These common intersections can also be on any small concentric circle, with the spacing of the measurement points on this circle being, for example, less than 0.5 mm, preferably less than 0.1 mm, particularly preferably less than 0.01 mm. be. Particularly preferably, the concentric ring has a diameter of less than 0.5 mm, preferably less than 0.1 mm, particularly preferably less than 0.01 mm. As a result, it is possible to achieve high resolution in the central region of the object in a simple way. This can be particularly advantageous in determining the surface morphology of the eye layer in ophthalmology.

The loop is preferably embodied so that three consecutive measurement points on the loop do not exist in a straight line. Depending on the density of measurement points, the reference can be extended to four or more consecutive measurement points that, in principle, do not exist on a straight line.

In the variant, the locus may also be non-exclusive, but may not be provided by the loop, and the three consecutive measurement points may be on a straight line. When the reference is extended to 4, 5, or 6 or more points, in the modified example, the locus portion of 4, 5, or 6 or more consecutive measurement points may exist on a straight line. Correspondingly possible. Adjacent loops do not necessarily have to intersect, or intersections do not necessarily have to be on a circular ring.

Preferably, the locus is defined by two frequencies and radii, and more specifically by exactly two frequencies. As a result, it becomes easier to calibrate the measurement method. It is possible to impose fewer requirements on frequency response for scanners.

In variants, it is also possible to use three or more frequencies or two or more radii to define the trajectory. The locus can also have a polygon or any other definition.

を有する曲線に追従する。ここで、
ｒ_０：スキャニングパターンの円周の半径
ω_Ｂ、ω_Ｔ：角速度
であり、ｒ_０は、測定される表面の円周の半径である。この関数は、軌跡であって、較正するのが特に簡単であり、また、正確に２つの周波数（ω_Ｂ及びω_Ｔ）及び１つの半径（ｒ_０）によって定義される、軌跡を構成する。この軌跡の更なる利点は、半径方向における増分が外側に向かって減少し、したがって、信号コントラストが増加することにある。これは、眼を測定するときに特に有利である。その理由は、眼の表面の勾配が、半径方向に眼の中心から離れると増加するからである。 Preferably, the measurement beam has the following coordinates:

Follows a curve with. here,
r ₀ : Radius of the circumference of the scanning pattern ω _B , ω _T : Angular velocity, and r ₀ is the radius of the circumference of the surface to be measured. This function is a locus, which is particularly easy to calibrate and constitutes a locus defined _{exactly by two frequencies (ω B} and ω _T ) and one radius (r _0). A further advantage of this trajectory is that the radial increment decreases outward and therefore the signal contrast increases. This is especially advantageous when measuring the eye. The reason is that the gradient of the surface of the eye increases radially away from the center of the eye.

In a particularly preferred embodiment, ω _B = 4 ^* 2π / t _pattern and ω _T = 7 ^* 2π / t _pattern , where t _pattern indicates the duration of the measurement cycle. However, the frequency may be selected differently depending on the size of the area to be measured in the case of an invariant measurement point density or depending on the desired measurement point density in the case of a constant size area. You can also. For example, in the case of a relatively small area or a relatively small measurement point density, ω _B = 4 ^* 2π / t _pattern and ω _T = 5 ^* 2π / t _pattern , and a relatively large area or a relatively large measurement point. In the case of density, ω _B = 4 ^* 2π / t _pattern and ω _T = 11 ^* 2π / t _pattern . It will be apparent to those skilled in the art that this trajectory can be parameterized as desired. In particular, it is also possible to select a non-integer multiple of 2π / t _{pattern for the angular frequency.} It is also apparent to those skilled in the art that the measurement beam does not need to follow this function in a mathematically exact manner, only the measurement point needs to be on the trajectory. Moreover, the measurement points do not have to be exactly on the trajectory, preferably, for example, less than 5%, preferably less than 1%, particularly preferably less than 0.1% of the diameter of the surface to be measured. It can only exist close enough to the trajectory with a distance of.

ここで、測定される表面の円周の半径はａ−ｂ＋ｃで与えられる。 In the variant, it is also possible to provide different functions to define the trajectory. As an example, the locus can be provided by the following inner trochoid coordinate equation.

Here, the radius of the circumference of the surface to be measured is given by ab + c.

In this case, the inner trochoid is formed by a radial course of arbitrary length on the inner circle, with the inner circle rolling within a larger circle. Therefore, a indicates the radius of the outer circle, and b indicates the radius of the inner circle having a pointer of length c. As a result, the radius of the circumference of the surface to be measured is given by ab + c.

In the inner trochoidal form, with proper parameterization, the center may remain open within a region. This is the case when b + c <a. Therefore, the radius of the inner opening circle (if present) is a− (b + c). As a result, it is possible to reach many different intersections in the central region of the circumference, and as a result, the measurement accuracy of the central region can be improved once again. The radius of the inner circle is preferably maintained so that coverage criteria are met. That is, the radius can be, for example, 0.25 mm or less.

The radius of the central opening region is exactly zero when ab = c, and as a result, multiple intersections are in the center. This form is preferred when the monitoring or correction of eye movement is weighted higher than the high resolution in the central region of the eye.

Those skilled in the art will notice any further deformation of the trajectory, which will also only have a large radius of curvature and will cover both the center and edge regions with high resolution.

Preferably, for each measurement point on the locus, there is a second measurement point on the same locus at a distance of less than 25%, preferably less than 16% of the radius of the circumference. Particularly preferably, on each circle of the measured surface, having a radius of less than 25%, preferably less than 16%, particularly preferably less than 5% of the circumference. There is at least one measurement point. As a result, an ideal measurement point density with sufficient resolution of the object is obtained.

In the modified example, the distance to the next measurement point may be limited differently.

Preferably, for each point within the circumference of the surface, there is one measurement point on the locus at a distance of at most 0.5 mm, preferably at most 0.25 mm, and particularly preferably at most 0.1 mm. As a result, it is possible to obtain ideal coverage with multiple measurement points on the surface being measured, especially when measuring the eye.

In a variant, coverage with multiple measurement points on the surface to be measured can also be defined differently. As an example, a circular disc with a radius of 0.5 mm, preferably 0.25 mm, particularly preferably 0.1 mm, which remains within the surface to be measured, always has at least one measurement point, regardless of the position of the circular disc. Further preferred deformations can be provided to preserve the distribution of measurement points on the surface to be measured so as to cover.

The present invention further relates to a method for approximating a cross section of an object. The measurement points required for the approximation method are preferably implemented using the methods described above for measuring an object.

However, it is also conceivable to use different methods and establish measurement points, in particular by conventional measurement methods known to those of skill in the art.

Therefore, in order to approximate the cross section of an object, preferably the eye, to approximate the cross section of a subset of recorded measurement points having at least one measurement point at a distance from the cross section in the area of the cross section of the eye. , The calculation performed on the recorded measurement points is performed. As a result, it is possible to calculate any cross section of an object without the need for each appropriate measurement point to be exactly available for the cross section.

Preferably, a subset of recorded measurement points of two sectors with a center-point angle of less than 90 degrees, arranged in a mirror symmetry manner, is the circumference of the surface to be measured. Is inside.

Therefore, in the case of the measurement method using a constant angular velocity, it is possible to achieve a substantially constant number of measurement points for one cross-sectional length portion over the cross section.

In the variant, the sectors are not similarly separated by straight lines, but rather cannot extend linearly from the center of the circumference in the radial direction. The weighting that can be applied thereby is that measurement points located far from the cross section provide greater unsharpness in approximation. As a result, more measurement points are included in these areas. It will be apparent to those skilled in the art that, in principle, any surface formed differently can be used to approximate the cross section. In particular, the cross section does not have to extend in a linear fashion as well, but rather can be embodied in a V-shaped manner, for example, as a "piece of cake" of the object, thereby. The range of influence is correspondingly two circular sectors, comprising two circular sectors that optionally have an intermediate angle or that touch or overlap each other. The cross section can also have different profiles, such as a wavy profile, such as a polygonal profile such as a regular polygon, a zigzag profile, and the like.

A combination of further advantageous embodiments and features of the present invention will become apparent from the following detailed description and the entire scope of claims.

The raster shape scanning pattern by the prior art is shown. The loop shape scanning pattern by the prior art is shown. The spiral shape scanning pattern by the prior art is shown. A first embodiment of a scanning pattern according to the present invention is shown. A second embodiment of the scanning pattern according to the present invention is shown. A third embodiment of the scanning pattern according to the present invention is shown. A fourth embodiment of the scanning pattern according to the present invention is shown.

In principle, the same parts have the same reference numerals in the figure.

Method for Carrying Out The Invention FIG. 1 shows a raster-shaped scanning pattern according to the prior art. The scanning pattern is known, among other things, from tube televisions through which electron beams pass line by line on the screen. FIG. 1 shows two equal patterns that are offset by 90 degrees with respect to each other and overlap each other. Both patterns move separately. Parts with straight trajectories can be traced at relatively high speeds, but each has a significant delay as a result of abrupt directional changes in the edge region. Therefore, this scanning pattern cannot be traced in a time-efficient manner by the measurement beam. Further, the scanning pattern has the same measurement point density in the edge region as the center, especially in the case of a circular or spherical crown shape such as an eye. Therefore, this scanning pattern or this trajectory does not provide an ideal embodiment for recording the surface of the eye.

FIG. 2 shows a loop-shaped scanning pattern according to the prior art. This scanning pattern has a plurality of loops formed from the center. In other words, the scanning pattern is provided by a straight portion, which passes over the center of the circumference of the measured surface at a constant angular distance. At each end of such a straight portion, for example, by looping clockwise, there is a connection to the adjacent portion. The loop ends formed have a relatively small radius of curvature. The greater the angular distance of the part, the greater the radius of curvature at the edge of the part, while the total number of parts decreases at the same time, and as a result, the density of measurement points on the surface being measured also decreases. As a result, the two factors of radius of curvature and measurement point density or maximum distance between two adjacent points compete with each other. A further drawback is that there is a very large increase in the density of measurement points in the center, which is not meaningfully useful in assessing the surface profile of the object. Optionally, the scanning pattern can be selected such that the majority of the scanning pattern is large enough to be outside the surface to be measured. The advantage is that the radius of curvature is larger and can be passed faster, while this will significantly increase the total path length and, overall, increase the measurement time. Become.

FIG. 3 shows a spiral shape scanning pattern according to the prior art. The spiral pattern in the edge region has a sufficiently large radius of curvature that can be passed at high speed by the measurement beam, but the radius of curvature becomes smaller and smaller towards the center. However, because the central region is so important, especially in ophthalmology, this scanning pattern also points that the central region can only pass very slowly or can only be measured with low resolution. It is disadvantageous.

を有する本発明による４つの異なるスキャニングパターンが示される。ここで、
ｒ_０：スキャニングパターンの円周の半径
ω_Ｂ：

ω_Ｔ：

である。 In the following FIGS. 4 to 7, the general type

Four different scanning patterns according to the invention are shown. here,
r ₀ : Radius of the circumference of the scanning pattern ω _B :

ω _T :

Is.

In the following example, the measurement time t _pattern is 200 ms (milliseconds). In principle, it will be apparent to those skilled in the art that the shortest possible measurement time is required. However, the shortest measurement time depends first on the measuring instrument used and then on the number of measuring points.

In this case, the number of measurement points is 3200, and the measurement frequency (that is, the rate at which the measurement points are recorded) is f = 16 kHz. Here, a balance is required that the measurement time is sufficiently short and, at the same time, the number of measurement points, and therefore the resolution is sufficiently large when a constant surface is measured. However, in addition, the measurement frequency is only high enough that sufficient signal strength still appears for each measurement point. The reason is that the signal strength decreases as the measurement frequency increases. Depending on the measurement system, the measurement frequency can be from a few kHz to a few MHz. Measurement frequencies in the range of 10 kHz to 200 kHz have been found to be valuable.

Depending on the measuring equipment used, the measurement time and the number of measurement points can also be less or more. Depending on the measurement arrangement configuration, it may be advantageous to reduce the measurement time, when lower resolution is accepted. On the other hand, it is also possible to increase the number of measurement points by impairing the measurement time.

In this case, the measured surface radius is 4 mm. However, the radius of the surface to be measured also depends on the particular requirements and, in principle, is optionally, for example, 10 mm, 3.5 mm, 1.5 mm, and between those values and their values. It can be said that it exists in the entire range that exists outside of.

The axial system resolution of the measuring instrument is in this case about 4.6 μm, but the resolution can be similarly higher or lower.

It will be apparent to those skilled in the art that the diameter, the number of measurement points, and the measurement time can be in different ranges.

Finally, it will be apparent to those skilled in the art that the locus is not limited to exactly matching the graph formed by the equations specified in the general form (the equations above and below). Trajectories or scanning patterns can also deviate from mathematically accurate forms. So, for example, the set of points established by the measurement beam can simply approximate such a function as an interpolation.

FIG. 4 shows a first embodiment of a scanning pattern according to the invention, which is a particularly preferred embodiment having B = 8 and T = 7. From the graph of the function, it is easy to identify that the radius of curvature increases from the edge region to the center, respectively. Further, each of the eight intersections is always present on a concentric circle at the center of the circumference, and the center point is passed many times. Furthermore, it is possible to see from the figure that both the edge region and the region near the center can be measured with high resolution. The scanning pattern has 48 simple intersections and one centered eight intersections. Eye movements can be detected and removed, especially by intersections away from the center. The large number of intersections allows the detection of eye movements with correspondingly high frequencies (measurement time / number of intersections = average update time).

FIG. 5 shows a second embodiment of the scanning pattern according to the invention, where B = 8 and T = 11. In contrast to the scanning pattern according to FIG. 4, this scanning pattern has a longer path length and more intersections on the same surface. This allows for higher resolution, i.e., shorter average distances between adjacent measurement points. The number of simple intersections is 80 in this case, and in each case has eight intersections for one concentric ring.

FIG. 6 shows a third embodiment of the scanning pattern according to the invention, where B = 13 and T = 14. With 26 intersections and 13 concentric rings for one concentric ring, the scanning pattern of this embodiment has a total of 338 intersections. This embodiment, or otherwise having more intersections, can be used in the case of a reasonably fast scanner, or for objects in a relatively large area. However, the measurement time along these trajectories is probably too long using current OCT scanners.

を有する。 Finally, as a fourth example, FIG. 7 shows an inner trochoidal scanning pattern as a possible embodiment of the scanning pattern according to the present invention. The inner trochoid scanning pattern is a general form

Have.

However, it will be apparent to those skilled in the art that, in this case, the trajectory is not limited to exactly following the graph formed by the above equations as well. Trajectories or scanning patterns can also deviate from mathematically accurate forms. So, for example, the set of points established by the measurement beam can simply approximate such a function as an interpolation.

To establish a measurement in ophthalmology, the value can be selected again to obtain a radius of about 4 mm. As an example, a = 2, b = 0.1, and c = 2.1 are selected in FIG. Using this parameterization, it is possible to identify a free circle with a radius of about 0.2 mm at the center of the circumference. Therefore, this free surface meets the 0.5 mm criterion mentioned at the beginning.

In summary, the method according to the invention for recording measurement points can be carried out particularly fast, and thus the method is robust to the movement of objects, especially the eyes, while at the same time, especially the spherical cap. It should be noted that high resolution is achievable in the edge region of the geometry.

Claims (38)

A method for recording a measurement point on the eye, the the controlled measurement beam along a trajectory on the curved surface of the eye, measurement points for recording the axial length profile is recorded, a method, The minimum radius of curvature of the locus is at least 1/7 of the radius of the circumference of the surface, and the measurement beam is the coordinate.

Where r ₀ is the radius of the circumference of the surface to be measured, so that the average deviation of the measurement points from the locus is 5 of the diameter of the circumference of the surface. The method, which is less than%. The method of claim 1, wherein at least 90% of the locus extends within the circumference. The method according to claim 1 or 2, wherein the radius of curvature of the locus is less than the radius of the circumference of the surface. The method according to any one of claims 1 to 3, wherein the maximum radius of curvature of the locus is less than 99% of the radius of the circumference. The method according to any one of claims 1 to 4, wherein the curvature of the locus toward the center of the circumference increases monotonically. The method of any one of claims 1-5, wherein the locus has intersections that are recorded at least twice with a time delay. The method of claim 6, wherein the locus has at least two spaced intersections. The method of claim 7, wherein the at least two intersections have an intersection angle greater than 90 degrees in plane projection. The measurement beam follows a locus having three or more intersections, and in the case of k * n intersections, each n intersections are present on one concentric ring of each of the k concentric rings. , The method according to claim 7 or 8. The method of claim 9, wherein the distance between two adjacent concentric rings with increasing radii is reduced between the three said concentric rings with the largest radius. The method according to claim 9 or 10, wherein the intersections between the locus and every other concentric ring exist on a straight line oriented in the radial direction, respectively. The method according to any one of claims 1 to 11, wherein the measurement beam is displaced along the projection of the trajectory at a constant angular velocity. The method according to any one of claims 1 to 12, wherein the measurement point is recorded by the spectral region OCT or the sweep light source OCT. The method according to any one of claims 1 to 13, wherein the locus is provided by a loop and adjacent loops intersect. The method according to any one of claims 1 to 14, wherein for each measurement point on the locus, a second measurement point on the same locus exists at a distance of less than 25% of the radius. The method according to any one of claims 1 to 15, wherein the measurement points on the locus exist at a distance of 0.5 mm at the longest for each point in the circumference of the surface. A method of approximating a cross section of an eye using the measurement points recorded by the method according to any one of claims 1 to 16, wherein in the region of the cross section, at least one at a distance from the cross section. A method in which operations performed on the recorded measurement points are performed to approximate the cross section of a subset of the recorded measurement points, including one measurement point. A device that implements the method according to any one of claims 1 to 16. The method according to claim 1, wherein the minimum radius of curvature of the locus is at least 1/5 of the radius of the circumference of the surface. 19. The method of claim 19, wherein the minimum radius of curvature of the locus is at least 1/3 of the radius of the circumference of the surface. 15. The method of claim 15, wherein the distance is less than 16% of the radius. The method according to claim 16, wherein the measurement points on the locus exist at a distance of 0.25 mm at the longest for each point in the circumference of the surface. The method according to claim 22, wherein the measurement points on the locus exist at a distance of 0.1 mm at the longest for each point in the circumference of the surface. The method of claim 2, wherein at least 95% of the locus extends within the circumference. 24. The method of claim 24, wherein the entire trajectory extends within the circumference. The method of claim 3, wherein the radius of curvature of the locus is less than the radius of the circumference of the surface over a path length greater than 50% of the locus. 26. The method of claim 26, wherein the radius of curvature of the locus is less than the radius of the circumference of the surface over a path length greater than 75% of the locus. 27. The method of claim 27, wherein the radius of curvature of the locus is less than the radius of the circumference of the surface over the entire path length of the locus. The method of claim 4, wherein the maximum radius of curvature is less than 99% of the radius of the circumference. 29. The method of claim 29, wherein the maximum radius of curvature is less than 95% of the radius of the circumference. 30. The method of claim 30, wherein the maximum radius of curvature is less than 90% of the radius of the circumference. The method of claim 5, wherein the curvature of the locus towards the center of the circumference increases strictly monotonously. The method of claim 6, wherein the locus has intersections that are recorded at least twice with a time delay, and as a result, the movement of the eye is detected. The method of claim 7, wherein the locus has at least two spaced intersections, the intersections being recorded three or more times with a time delay. 13. The method of claim 13, wherein the measurement points are recorded at a constant frequency over time. 14. The method of claim 14, wherein the two adjacent loops each have intersections that lie on a circle. 36. The method of claim 36, wherein the intersections are on circles concentric with the circumference of the surface to be measured. The method of claim 1, wherein the locus is defined exactly by two frequencies and a radius.

Images